I’m ashamed to admit that I once suddenly realized what I had done: I asked the LLM to rename a variable. It took it some minutes or so. With IDE by hand it would have been seconds.
The other day I asked it to update a handful of files using a new pattern I built. It was tedious work, and probably would have taken me 10 minutes, but I figure why not just let Claude do it. Well, it spun up multiple agents, each taking a different approach (one writing a Perl script, the other just doing fancy regex, etc) and 30 minutes later had the result. Granted, it wasn't my 30 minutes it was using, but it felt wasteful nonetheless.
Now, if it's something minor, I'll ask Claude about it, and get the gist of what I need to do, and just do it myself.
The annoying issue sometimes is that while it is faster to do it yourself, the context of the LLM is not updated. So now you have you to mention that and mentioning it is a bit fuzzier sometimes than the LLM simply doing the rename itself, so the context stays up to date.
Honest take: I work less now with maybe slightly faster results while the output quality is probably about the same. Some bugs sometimes fall through, but when hasn’t that been the case?
I take effort to write specs for the agent and I take effort reading PR’s but thats maybe an hour of work. Obviously there are meetings, sometimes tasks require more planning, sometimes I investigate production problems/data more manually, but overall - I feel I work less after agents became a thing. Kind of ashamed of it tbh.
I feel the 1M context is way too large —- the model gets ”drunk” way before it gets anywhere near 1 million. Imo the 1M context window is a huge downgrade.
I suspect people with opinions like the author’s haven’t been in a project where people use LLMs responsibly. We had a senior dev basically just prompt and push, with very little overseeing and minimal instructions, causing so many bad PRs and even prod bugs. That made me a sceptic for many months about agents.
Then we started to have (myself included) people actual plan out the tasks for the bot: give good specs, good ac, file context, better self-review, better ”agentic practices” (i.e. asking it to review it own work can sometimes help), and suddenly I noticed you really can use agents in a real world 1mil LOC project. If you do it well and responsibly (also meaning you still retain some sense of ownership and actually review the shit)
Nope. My point is not about the quality of the generated code, it's the fact that it was generated. No matter how good it is it will always be a liability without the accompanying asset which is the understanding produced by undertaking the effort of writing the code. Generated code is exclusively cognitive debt. It is also, by definition, legacy code since no one wrote it.
Look I hate AI "prose" and "art" with a deep passion, and i refuse to let an LLM touch my documentation, but this idea that you can't use an LLM to write decent code that you genuinely comprehend is outdated. Many developers choose not to, but that doesn't mean it's not possible.
I have coworkers who use an LLM who can explain and defend every line of every PR. I also have coworkers who just prompt and pray. Only one camp is producing genuine liabilities.
I use LLM’s daily and agents occasionally. They are useful, but there is no need to move any goal posts; they easily do shit work still in 2026.
All my coworkers use agents extensively in the backend and the amount of shit code, bad tests and bugs has skyrocketed.
Couple that with a domain (medicine) where our customer in some cases needs to validate the application’s behaviour extensively and it’s a fucking disaster —- very expensive iteration instead of doing it well upfront.
I think we have some pretty good power tools now, but using them appropriately is a skill issue, and some people are learning to use them in a very expensive way.
I find that chat is pretty good when you're describing what you want to do, for saying "actually, I wanted something different," or for giving it a bug report. For making fine adjustments to CSS, it would be nice if you could ask the bot for a slider or a color picker that makes live updates.
I just dont understand how anyone can go "yeah okay lets install these guys rootkit complete with keylogger and who knows what, totally legit!", and for what? to play a game
What I’ve seen also happen is senior devs suddenly starting to put out garbage code and PRs. One senior dev in our project has become a menace and the quality of his work has dramatically dropped.
I graduated literally 3 months ago so that's my skill level.
I also have no idea what the social norms are for AI. I posted the comment after a friend on Discord said I should disclose my use of AI.
The underlying purpose of the PR is ironically because Cline and Copilot keep trying to use `int` when modern C++ coding standards suggest `size_t` (or something similar).
